Application Prospects of Metabolic P System / Metabolinės P Sistemos Taikymo Galimybės

Abstract

Metabolic P (MP) systems are a class of P systems and are based on metabolic process which is used by living organisms. We review for the purpose of experimental research the MP system development, implementation and modeling. To implement MP systems in dedicated hardware, some specifications need to be determined. We look into the possibility of modeling MP systems in fixed point arithmetic capable hardware and determine the minimal needed bit count for selected MP system model. A way to improve the accuracy of fixed point calculations is proposed, which includes the modification of MP grammar. This modification is done using a simple rule to multiply and divide constants in MP grammar equations. The resulting calculations have more than 250 times smaller error when compared to original calculations. When investigating the performance of MP calculations, it is determined that the time needed for all calculations to finish depends linearly on the number of modeled systems. Metabolinė P sistema yra P sistemos klasė, pagrįsta metabolinio proceso, būdingo gyviesiems organizmams, taikymu. Siekiant pasirengti įgyvendinti MP sistemas aparatine įranga, pateikiama MP sistemų kūrimo, įgyvendinimo ir modeliavimo literatūros analitinė apžvalga. Įvertinama galimybė modeliuoti MP sistemas taikant fiksuotojo kablelio aritmetiką. Imitaciniais eksperimentais nustatytas minimalus pasirinktam MP sistemos modeliui taikyti reikiamas žodžių ilgis. Pasiūlytas būdas fiksuotosios aritmetikos skaičiavimo tikslumui padidinti. Šio būdo esmė yra MP gramatikos modifikavimas pagal konstantų daugybos ir dalybos taisyklę. Skaičiavimus atliekant modifikuota sistema, paklaida esti daugiau nei 250 kartų mažesnė. Tiriant MP sistemų našumą nustatyta, kad skaičiavimo trukmė tiesiškai priklauso nuo modeliuojamų sistemų skaičiaus.

Research of Metabolic P System Field Programmable Gate Array Implementation

“…The research results are published in author publication (Kulakovskis 2015). The main results are announced in national "Science -Future of Lithuania" (Vilnius, 2015) scientific conference.…”

Automated Metabolic P System Placement in FPGA

“…MATLAB software was chosen because it was already used in previous experiments [14]. The implementation in software can be used as a reference point as it uses the more precise floating-point arithmetic than the fixed-point FPGA implementation.…”

“…Although MP systems can be implemented in software using specialized MetaPlab software or relatively simple implementation of MP formulae in MATLAB [14], it is also possible to directly implement them in hardware. P systems are known to have been implemented in hardware [15], [16], but so far there are no known similar implementations of MP systems.…”